Homer Closest TSS per Gene Data and Heatmaps¶

2-6-26 Update

In [1]:
import pandas as pd
import numpy as np
In [2]:
cd GEO/RNA-seq/pca_ma_plots/final/

/vf/users/gallegosda/GEO/RNA-seq/pca_ma_plots/final
In [3]:
homerClosestTSSperGene_df = pd.read_csv("homer.closestTSS_perGene.csv", names=['geneName','var2','var3','var4','var5','var6','var7','var8','var9','var10','var11','var12','var13','var14','var15','var16','var17','var18','var19'])
In [4]:
homerClosestTSSperGene_df.head(3)
Out[4]:
geneName var2 var3 var4 var5 var6 var7 var8 var9 var10 var11 var12 var13 var14 var15 var16 var17 var18 var19
0 Car4 AW456718|Ca4 carbonic anhydrase 4 443 chr11 84957788 84958215 + 0 NaN intron (NM_007607, intron 1 of 7) CpG 247 NM_007607 12351 Mm.1641 NM_007607 ENSMUSG00000000805 protein-coding
1 Ppp3ca 2900074D19Rik|CN|Caln|Calna|CnA protein phosphatase 3, catalytic subunit, alph... 1885 chr3 136670472 136671032 + 0 NaN 5' UTR (NM_008913, exon 1 of 14) 5' UTR (NM_008913, exon 1 of 14) 686 NM_008913 19055 Mm.331389 NM_008913 ENSMUSG00000028161 protein-coding
2 Nr5a1 Ad4BP|ELP|ELP-3|Ftz-F1|Ftzf1|SF-1|SF1|STF-1 nuclear receptor subfamily 5, group A, member 1 1550 chr2 38711554 38711989 + 0 NaN intron (NM_139051, intron 1 of 6) CpG-8071 463 NM_001316687 26423 Mm.31387 NM_139051 ENSMUSG00000026751 protein-coding
In [5]:
allMouseGenes_df = pd.read_csv("all_mouse_gene_ENSEMBL_IDs_and_gene_names.txt",names=['ensemblID','stableID','stableTranscript','transcriptV','geneName'],skiprows=1)
In [6]:
allMouseGenes_df
Out[6]:
ensemblID stableID stableTranscript transcriptV geneName
0 ENSMUSG00000064336 ENSMUSG00000064336.1 ENSMUST00000082387 ENSMUST00000082387.1 mt-Tf
1 ENSMUSG00000064337 ENSMUSG00000064337.1 ENSMUST00000082388 ENSMUST00000082388.1 mt-Rnr1
2 ENSMUSG00000064338 ENSMUSG00000064338.1 ENSMUST00000082389 ENSMUST00000082389.1 mt-Tv
3 ENSMUSG00000064339 ENSMUSG00000064339.1 ENSMUST00000082390 ENSMUST00000082390.1 mt-Rnr2
4 ENSMUSG00000064340 ENSMUSG00000064340.1 ENSMUST00000082391 ENSMUST00000082391.1 mt-Tl1
... ... ... ... ... ...
278391 ENSMUSG00000026833 ENSMUSG00000026833.19 ENSMUST00000152415 ENSMUST00000152415.2 Olfm1
278392 ENSMUSG00000026833 ENSMUSG00000026833.19 ENSMUST00000113920 ENSMUST00000113920.8 Olfm1
278393 ENSMUSG00000026833 ENSMUSG00000026833.19 ENSMUST00000100244 ENSMUST00000100244.10 Olfm1
278394 ENSMUSG00000026833 ENSMUSG00000026833.19 ENSMUST00000102879 ENSMUST00000102879.4 Olfm1
278395 ENSMUSG00000026833 ENSMUSG00000026833.19 ENSMUST00000028177 ENSMUST00000028177.11 Olfm1

278396 rows × 5 columns

In [7]:
# This function returns a df from a file like TEtranscripts_GRCm38_E10_777tm1d_2KO_males_vs_1WT_female_1WT_male_non_stranded_gene_TE_analysis.txt
# Receives .txt as input and returns df

def getGeneNameForTEtranscripts_txt(TEtranscripts_txtFile):
    df = pd.read_csv(TEtranscripts_txtFile, sep='\t',index_col=False,names=['ensemblID','baseMean','log2FoldChange','lfcSE','stat','pvalue','padj'],skiprows=1)
    
    df["ensemblID"] = (
        df["ensemblID"]
        .astype("string")          # pandas StringDtype (not plain str)
        .str.replace(r"\..*", "", regex=True)
    ) 
    
    df = df.merge(
        allMouseGenes_df[["geneName", "ensemblID"]],
        on="ensemblID",
        how="left"
    )
    
    df = df.drop_duplicates()

    return df
In [8]:
# df = pd.read_csv("TEtranscripts_GRCm38_E10_777tm1d_2KO_males_vs_1WT_female_1WT_male_non_stranded_gene_TE_analysis.txt", sep='\t',index_col=False,names=['ensemblID','baseMean','log2FoldChange','lfcSE','stat','pvalue','padj'],skiprows=1)
# # print(df.head)

# # Rename the first column (at index 0)
# # df = df.rename(columns={df.columns[0]: 'ensemblID'})

# df["ensemblID"] = (
#     df["ensemblID"]
#     .astype("string")          # pandas StringDtype (not plain str)
#     .str.replace(r"\..*", "", regex=True)
# )

# # print(df.head)
# # df

# df = df.merge(
#     allMouseGenes_df[["geneName", "ensemblID"]],
#     on="ensemblID",
#     how="left"
# )

# df = df.drop_duplicates()

# df
In [9]:
mMEF_E15_aKOvWT_males_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_mMEF_E15_777KO_tm1a_vs_WT_males_non_stranded_gene_TE_analysis.txt")
In [10]:
mMEF_E15_aKOvWT_females_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_mMEF_E15_777KO_tm1a_vs_WT_females_non_stranded_gene_TE_analysis.txt")
In [11]:
mMEF_E15_patientKIvE16WT_females_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_mMEF_E15_777-R297W-KI_vs_E16_WT_females_non_stranded_gene_TE_analysis.txt")
In [12]:
mMEF_E15_patientKIvE15WT_males_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_mMEF_E15_777-R297W-KI_vs_E15_WT_males_non_stranded_gene_TE_analysis.txt")
In [13]:
mMEF_E13_clusterKOvWT_males_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_mMEF_E13_DUFKZFP_cluster_KO_vs_WT_males_non_stranded_gene_TE_analysis.txt")
In [14]:
mF9_OE_patientpCMV6vpSBmock_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_mF9_OE_pCMV6_777-R297W-HA_vs_pSB_mock_non_stranded_gene_TE_analysis.txt")
In [15]:
mF9_OE_777vpSBmock_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_mF9_OE_pCMV6_777-HA_vs_pSB_mock_non_stranded_gene_TE_analysis.txt")
In [16]:
mESC_KO_EGFPexcisedvWT_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_mESC_R1_777KO_EGFP_excised_vs_WT_non_stranded_gene_TE_analysis.txt")
In [17]:
E8_aKOvWT_males_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_E8_777tm1a_KO_vs_WT_males_non_stranded_gene_TE_analysis.txt")
In [18]:
E8_aKOvWT_females_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_E8_777tm1a_KO_vs_WT_females_non_stranded_gene_TE_analysis.txt")
In [19]:
E10_dKOvWT_mf_df = getGeneNameForTEtranscripts_txt("TEtranscripts_GRCm38_E10_777tm1d_2KO_males_vs_1WT_female_1WT_male_non_stranded_gene_TE_analysis.txt")
In [ ]:
In [20]:
mMEF_E15_aKOvWT_males_df.loc[mMEF_E15_aKOvWT_males_df['geneName']=='Acadl']
Out[20]:
ensemblID baseMean log2FoldChange lfcSE stat pvalue padj geneName
29030 ENSMUSG00000026003 1309.852514 0.029923 0.222277 0.13462 0.892913 0.981316 Acadl
In [21]:
mF9_OE_patientpCMV6vpSBmock_df.loc[mF9_OE_patientpCMV6vpSBmock_df['geneName']=='Acadl']
Out[21]:
ensemblID baseMean log2FoldChange lfcSE stat pvalue padj geneName
28735 ENSMUSG00000026003 3563.687259 -0.059661 0.074049 -0.805698 0.420417 0.999695 Acadl
In [22]:
mMEF_E15_patientKIvE16WT_females_df.loc[mMEF_E15_patientKIvE16WT_females_df['geneName']=='Morc4']
Out[22]:
ensemblID baseMean log2FoldChange lfcSE stat pvalue padj geneName
46493 ENSMUSG00000031434 840.227535 -1.272542 0.2929 -4.344633 0.000014 0.000212 Morc4
In [23]:
mMEF_E13_clusterKOvWT_males_df.loc[mMEF_E13_clusterKOvWT_males_df['geneName']=='Lrrc7']
Out[23]:
ensemblID baseMean log2FoldChange lfcSE stat pvalue padj geneName
36227 ENSMUSG00000028176 7.217716 0.354474 2.071381 0.171129 0.864122 0.999969 Lrrc7
120807 ENSMUSG00000104597 12.254025 -0.081791 1.531621 -0.053402 0.957412 0.999969 Lrrc7
In [ ]:
In [24]:
mMEF_E15_aKOvWT_males_df
Out[24]:
ensemblID baseMean log2FoldChange lfcSE stat pvalue padj geneName
0 ENSMUSG00000000001 3123.590474 -0.012997 0.225714 -0.057583 9.540809e-01 9.836959e-01 Gnai3
1 ENSMUSG00000000028 350.340286 0.305006 0.279101 1.092816 2.744747e-01 7.623687e-01 Cdc45
5 ENSMUSG00000000031 21314.408858 -2.401234 0.274553 -8.745967 2.211141e-18 2.637364e-15 H19
26 ENSMUSG00000000037 19.745926 -1.137889 0.799598 -1.423075 1.547143e-01 6.026881e-01 Scml2
35 ENSMUSG00000000049 39.011157 -0.248033 0.586197 -0.423122 6.722060e-01 9.259562e-01 Apoh
... ... ... ... ... ... ... ... ...
128928 ZP3AR:Satellite:Satellite 626.975354 -0.610377 0.323508 -1.886742 5.919505e-02 3.684686e-01 NaN
128929 Zaphod2:hAT-Tip100:DNA 0.912345 3.191318 3.832658 0.832664 4.050341e-01 8.648981e-01 NaN
128930 Zaphod3:hAT-Tip100:DNA 16.753550 -0.781496 0.890346 -0.877744 3.800825e-01 8.518641e-01 NaN
128931 Zaphod:hAT-Tip100:DNA 13.651162 -0.571859 0.995754 -0.574297 5.657667e-01 9.105791e-01 NaN
128932 hAT-N1_Mam:hAT:DNA 10.122079 -0.684578 1.095539 -0.624878 5.320514e-01 9.105791e-01 NaN

25048 rows × 8 columns

In [25]:
homerClosestTSSperGene_GeneNamesOnly_df = homerClosestTSSperGene_df[['geneName']]
In [26]:
# Test join
df = homerClosestTSSperGene_GeneNamesOnly_df.merge(
    mMEF_E15_aKOvWT_males_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
In [27]:
# Rename a specific column 'B' to 'New_B'
df = df.rename(columns={'log2FoldChange': 'mMEF_E15_aKOvWT_males_log2FC'})
In [28]:
df
Out[28]:
geneName mMEF_E15_aKOvWT_males_log2FC
0 Car4 1.339315
1 Ppp3ca 0.406624
2 Nr5a1 -4.865477
3 Bbs5 0.998045
4 Ppp3cc 0.358830
... ... ...
2798 Rell1 0.260231
2799 Car2 -1.666589
2800 Efna2 -0.752565
2801 Rell2 0.210392
2802 Efna3 0.516740

2803 rows × 2 columns

In [29]:
# More joins
df = df.merge(
    mMEF_E15_aKOvWT_females_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'mMEF_E15_aKOvWT_females_log2FC'})

df = df.merge(
    mMEF_E15_patientKIvE16WT_females_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'mMEF_E15_patientKIvE16WT_females_log2FC'})


df = df.merge(
    mMEF_E15_patientKIvE15WT_males_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'mMEF_E15_patientKIvE15WT_males_log2FC'})


df = df.merge(
    mMEF_E13_clusterKOvWT_males_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'mMEF_E13_clusterKOvWT_males_log2FC'})


df = df.merge(
    mF9_OE_patientpCMV6vpSBmock_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'mF9_OE_patientpCMV6vpSBmock_log2FC'})

df = df.merge(
    mF9_OE_777vpSBmock_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'mF9_OE_777vpSBmock_log2FC'})

df = df.merge(
    mESC_KO_EGFPexcisedvWT_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'mESC_KO_EGFPexcisedvWT_log2FC'})

df = df.merge(
    E8_aKOvWT_males_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'E8_aKOvWT_males_log2FC'})

df = df.merge(
    E8_aKOvWT_females_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'E8_aKOvWT_females_log2FC'})
# 
df = df.merge(
    E10_dKOvWT_mf_df[["geneName", "log2FoldChange"]],
    on="geneName",
    how="left"
)
df = df.rename(columns={'log2FoldChange': 'E10_dKOvWT_mf_log2FC'})
In [30]:
df
Out[30]:
geneName mMEF_E15_aKOvWT_males_log2FC mMEF_E15_aKOvWT_females_log2FC mMEF_E15_patientKIvE16WT_females_log2FC mMEF_E15_patientKIvE15WT_males_log2FC mMEF_E13_clusterKOvWT_males_log2FC mF9_OE_patientpCMV6vpSBmock_log2FC mF9_OE_777vpSBmock_log2FC mESC_KO_EGFPexcisedvWT_log2FC E8_aKOvWT_males_log2FC E8_aKOvWT_females_log2FC E10_dKOvWT_mf_log2FC
0 Car4 1.339315 -1.011783 -3.589685 NaN -3.683206 -1.305721 -0.273210 -0.489246 0.524201 0.194931 0.375135
1 Ppp3ca 0.406624 -0.104624 0.001710 0.450459 -0.276674 -0.118119 -0.211948 0.231611 -0.149899 -0.124822 0.226837
2 Nr5a1 -4.865477 1.510755 -3.627988 -4.042709 1.480193 0.222983 0.401732 0.104631 0.309612 1.176672 -0.184091
3 Bbs5 0.998045 0.012541 0.765101 0.823408 0.262260 -0.027248 -0.039019 -0.415166 0.031988 0.147180 -0.056199
4 Ppp3cc 0.358830 0.194157 0.157946 0.003984 0.298265 -0.048729 -0.261857 0.509793 0.209369 0.113688 0.548918
... ... ... ... ... ... ... ... ... ... ... ... ...
11299 Rell1 0.260231 0.282502 0.037566 -0.301917 0.371903 -0.111070 -0.219812 -0.082160 -0.033426 -0.008482 0.019051
11300 Car2 -1.666589 0.676888 0.011434 0.868281 -0.657800 0.102580 0.150735 -0.158000 0.132282 -0.062851 0.284113
11301 Efna2 -0.752565 0.018711 -1.173163 -1.734658 0.266176 -0.028690 0.064358 0.048713 -0.273028 -0.010901 -0.299740
11302 Rell2 0.210392 0.180841 0.070738 -0.472328 0.469701 0.032110 -0.152993 -0.254215 0.218178 0.336738 0.453236
11303 Efna3 0.516740 0.057861 1.523769 0.957994 -0.837704 0.041005 0.292736 -0.330265 -0.341390 -0.029855 0.069505

11304 rows × 12 columns

In [31]:
df = df.drop_duplicates()
In [39]:
df.loc[df['geneName']=='Lrrc7']
Out[39]:
geneName mMEF_E15_aKOvWT_males_log2FC mMEF_E15_aKOvWT_females_log2FC mMEF_E15_patientKIvE16WT_females_log2FC mMEF_E15_patientKIvE15WT_males_log2FC mMEF_E13_clusterKOvWT_males_log2FC mF9_OE_patientpCMV6vpSBmock_log2FC mF9_OE_777vpSBmock_log2FC mESC_KO_EGFPexcisedvWT_log2FC E8_aKOvWT_males_log2FC E8_aKOvWT_females_log2FC E10_dKOvWT_mf_log2FC
409 Lrrc7 -1.919093 -0.668156 -0.992237 -1.742988 0.354474 0.663596 0.29079 0.226616 0.015112 -0.003619 1.194863
410 Lrrc7 -1.919093 -0.668156 -0.992237 -1.742988 0.354474 0.663596 0.29079 0.226616 0.015112 -0.003619 -0.354013
411 Lrrc7 -1.919093 -0.668156 -0.992237 -1.742988 0.354474 0.663596 0.29079 0.226616 0.015112 -0.126133 1.194863
412 Lrrc7 -1.919093 -0.668156 -0.992237 -1.742988 0.354474 0.663596 0.29079 0.226616 0.015112 -0.126133 -0.354013
413 Lrrc7 -1.919093 -0.668156 -0.992237 -1.742988 0.354474 0.663596 0.29079 0.226616 -0.112248 -0.003619 1.194863
... ... ... ... ... ... ... ... ... ... ... ... ...
660 Lrrc7 0.027423 -0.210362 0.229261 -0.060135 -0.081791 0.663596 0.29079 0.226616 0.015112 -0.126133 -0.354013
661 Lrrc7 0.027423 -0.210362 0.229261 -0.060135 -0.081791 0.663596 0.29079 0.226616 -0.112248 -0.003619 1.194863
662 Lrrc7 0.027423 -0.210362 0.229261 -0.060135 -0.081791 0.663596 0.29079 0.226616 -0.112248 -0.003619 -0.354013
663 Lrrc7 0.027423 -0.210362 0.229261 -0.060135 -0.081791 0.663596 0.29079 0.226616 -0.112248 -0.126133 1.194863
664 Lrrc7 0.027423 -0.210362 0.229261 -0.060135 -0.081791 0.663596 0.29079 0.226616 -0.112248 -0.126133 -0.354013

256 rows × 12 columns

In [33]:
# df.to_csv('homerClosestTSSperGene_allTEtranscripts_log2FC.csv',index=False)

Heatmaps¶

In [34]:
homerClosestTSSperGene_df
Out[34]:
geneName var2 var3 var4 var5 var6 var7 var8 var9 var10 var11 var12 var13 var14 var15 var16 var17 var18 var19
0 Car4 AW456718|Ca4 carbonic anhydrase 4 443 chr11 84957788 84958215 + 0 NaN intron (NM_007607, intron 1 of 7) CpG 247 NM_007607 12351 Mm.1641 NM_007607 ENSMUSG00000000805 protein-coding
1 Ppp3ca 2900074D19Rik|CN|Caln|Calna|CnA protein phosphatase 3, catalytic subunit, alph... 1885 chr3 136670472 136671032 + 0 NaN 5' UTR (NM_008913, exon 1 of 14) 5' UTR (NM_008913, exon 1 of 14) 686 NM_008913 19055 Mm.331389 NM_008913 ENSMUSG00000028161 protein-coding
2 Nr5a1 Ad4BP|ELP|ELP-3|Ftz-F1|Ftzf1|SF-1|SF1|STF-1 nuclear receptor subfamily 5, group A, member 1 1550 chr2 38711554 38711989 + 0 NaN intron (NM_139051, intron 1 of 6) CpG-8071 463 NM_001316687 26423 Mm.31387 NM_139051 ENSMUSG00000026751 protein-coding
3 Bbs5 1700049I01Rik|2700023J09Rik Bardet-Biedl syndrome 5 (human) 1569 chr2 69647067 69647652 + 0 NaN intron (NM_028284, intron 1 of 11) CpG 188 NM_001362706 72569 Mm.252136 NM_028284 ENSMUSG00000063145 protein-coding
4 Ppp3cc Calnc|PP2BA gamma protein phosphatase 3, catalytic subunit, gamm... 922 chr14 70289129 70289377 + 0 NaN 5' UTR (NM_001360229, exon 1 of 13) 5' UTR (NM_001360229, exon 1 of 13) 244 NM_001304992 19057 Mm.439683 NM_008915 ENSMUSG00000022092 protein-coding
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
2793 Rell1 AA536743 RELT-like 1 2181 chr5 63968509 63968817 + 0 NaN exon (NM_145923, exon 1 of 7) exon (NM_145923, exon 1 of 7) 234 NM_145923 100532 Mm.243632 NM_145923 ENSMUSG00000047881 protein-coding
2794 Car2 AI131712|CAII|Ca2|Car-2|Ltw-5|Lvtw-5 carbonic anhydrase 2 1743 chr3 14886255 14887104 + 0 NaN 5' UTR (NM_009801, exon 1 of 7) 5' UTR (NM_009801, exon 1 of 7) 391 NM_009801 12349 Mm.1186 NM_009801 ENSMUSG00000027562 protein-coding
2795 Efna2 CEK7L|Elf1|Epl6|Eplg6|Lerk6 ephrin A2 247 chr10 80190330 80190907 + 0 NaN TTS (NM_007909) TTS (NM_007909) 11136 NM_007909 13637 Mm.1478 NM_007909 ENSMUSG00000003070 protein-coding
2796 Rell2 4631403P03Rik|ependolin RELT-like 2 1341 chr18 37955007 37956344 + 0 NaN promoter-TSS (NM_010411) promoter-TSS (NM_010411) 116 NM_153793 225392 Mm.233516 NM_153793 ENSMUSG00000044024 protein-coding
2797 Efna3 AW494418|EFL-2|Ehk1-L|Epl3|LERK-3 ephrin A3 1813 chr3 89321744 89322243 + 0 NaN intron (NM_010108, intron 1 of 4) CpG 886 NM_010108 13638 Mm.331159 NM_010108 ENSMUSG00000028039 protein-coding

2798 rows × 19 columns

In [35]:
boolean = not homerClosestTSSperGene_df["geneName"].is_unique      # True (credit to @Carsten)
In [36]:
boolean
Out[36]:
False
In [40]:
# df = your dataframe

# 1) identify log2FC columns
log2fc_cols = [c for c in df.columns if c.endswith("_log2FC")]

# 2) define threshold for a "strong" effect
#    set to 0 if you want ANY non-zero effect to count
threshold = 0.0

# absolute values
abs_fc = df[log2fc_cols].abs()

# 3) scoring
df_scored = df.copy()
df_scored["_n_strong"] = (abs_fc > threshold).sum(axis=1)
df_scored["_sum_abs"] = abs_fc.sum(axis=1)
df_scored["_max_abs"] = abs_fc.max(axis=1)  # optional tie-breaker

# 4) pick best row per gene
df_best = (
    df_scored
    .sort_values(
        ["geneName", "_n_strong", "_sum_abs", "_max_abs"],
        ascending=[True, False, False, False]
    )
    .drop_duplicates("geneName", keep="first")
    .drop(columns=["_n_strong", "_sum_abs", "_max_abs"])
)

# df_best now has exactly one row per geneName
In [41]:
df_best
Out[41]:
geneName mMEF_E15_aKOvWT_males_log2FC mMEF_E15_aKOvWT_females_log2FC mMEF_E15_patientKIvE16WT_females_log2FC mMEF_E15_patientKIvE15WT_males_log2FC mMEF_E13_clusterKOvWT_males_log2FC mF9_OE_patientpCMV6vpSBmock_log2FC mF9_OE_777vpSBmock_log2FC mESC_KO_EGFPexcisedvWT_log2FC E8_aKOvWT_males_log2FC E8_aKOvWT_females_log2FC E10_dKOvWT_mf_log2FC
722 0610010K14Rik NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN
109 0610039H22Rik NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN
936 1-Mar NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN
6536 1010001N08Rik NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN
6639 11-Mar NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN
... ... ... ... ... ... ... ... ... ... ... ... ...
1285 Zswim6 -0.096579 -0.039212 0.521084 -0.052862 -0.474079 -0.035291 -0.238596 0.359563 -0.165467 -0.040583 0.131327
6943 Zxda 3.780557 -1.040589 -1.336613 0.625054 NaN -4.390570 0.865508 NaN -0.308997 -0.161825 -0.232630
6946 Zxdb -0.291974 -0.261189 -0.222524 -0.098669 -0.063877 -0.062832 -0.114814 0.251703 -0.065300 0.016925 -0.000288
6173 Zyg11a 1.550915 1.519862 -1.048086 0.019624 -0.378368 -0.039749 -0.157374 0.501974 1.434094 1.037042 1.620861
6390 Zzz3 -0.072402 0.145015 -0.490400 -0.188292 0.091953 0.056168 0.233457 0.225465 -0.071601 0.014532 0.031210

2798 rows × 12 columns

In [42]:
# make geneName an ordered categorical using the other dataframe
order = homerClosestTSSperGene_df["geneName"]

df_sorted = (
    df_best.assign(
        geneName=pd.Categorical(df_best["geneName"], categories=order, ordered=True)
    )
    .sort_values("geneName")
)
In [48]:
df_sorted
Out[48]:
geneName mMEF_E15_aKOvWT_males_log2FC mMEF_E15_aKOvWT_females_log2FC mMEF_E15_patientKIvE16WT_females_log2FC mMEF_E15_patientKIvE15WT_males_log2FC mMEF_E13_clusterKOvWT_males_log2FC mF9_OE_patientpCMV6vpSBmock_log2FC mF9_OE_777vpSBmock_log2FC mESC_KO_EGFPexcisedvWT_log2FC E8_aKOvWT_males_log2FC E8_aKOvWT_females_log2FC E10_dKOvWT_mf_log2FC
0 Car4 1.339315 -1.011783 -3.589685 NaN -3.683206 -1.305721 -0.273210 -0.489246 0.524201 0.194931 0.375135
1 Ppp3ca 0.406624 -0.104624 0.001710 0.450459 -0.276674 -0.118119 -0.211948 0.231611 -0.149899 -0.124822 0.226837
2 Nr5a1 -4.865477 1.510755 -3.627988 -4.042709 1.480193 0.222983 0.401732 0.104631 0.309612 1.176672 -0.184091
3 Bbs5 0.998045 0.012541 0.765101 0.823408 0.262260 -0.027248 -0.039019 -0.415166 0.031988 0.147180 -0.056199
4 Ppp3cc 0.358830 0.194157 0.157946 0.003984 0.298265 -0.048729 -0.261857 0.509793 0.209369 0.113688 0.548918
... ... ... ... ... ... ... ... ... ... ... ... ...
11299 Rell1 0.260231 0.282502 0.037566 -0.301917 0.371903 -0.111070 -0.219812 -0.082160 -0.033426 -0.008482 0.019051
11300 Car2 -1.666589 0.676888 0.011434 0.868281 -0.657800 0.102580 0.150735 -0.158000 0.132282 -0.062851 0.284113
11301 Efna2 -0.752565 0.018711 -1.173163 -1.734658 0.266176 -0.028690 0.064358 0.048713 -0.273028 -0.010901 -0.299740
11302 Rell2 0.210392 0.180841 0.070738 -0.472328 0.469701 0.032110 -0.152993 -0.254215 0.218178 0.336738 0.453236
11303 Efna3 0.516740 0.057861 1.523769 0.957994 -0.837704 0.041005 0.292736 -0.330265 -0.341390 -0.029855 0.069505

2798 rows × 12 columns

In [72]:
import plotly.graph_objects as go

log2fc_cols = [c for c in df.columns if c != "geneName"]

z = df[log2fc_cols].to_numpy()

fig = go.Figure(
    data=go.Heatmap(
        z=z,
        x=log2fc_cols,
        y=df["geneName"],
        zmid=0,  # center colors around 0
        hovertemplate="Gene=%{y}<br>Group=%{x}<br>log2FC=%{z}<extra></extra>",
        colorscale='RdBu'
    )
)

fig.update_yaxes(showticklabels=False)

fig.update_layout(
    xaxis_title="Experimental group",
    yaxis_title="homer.closestTSS_perGene.tsv",
    height=1000,
    width=500
)

fig.show()
In [74]:
fig = go.Figure(
    go.Heatmap(
        z=df[log2fc_cols].to_numpy(),
        x=log2fc_cols,
        y=df["geneName"],
        zmin=-1,
        zmax=1,
        zmid=0,
        hovertemplate="Gene=%{y}<br>Group=%{x}<br>log2FC=%{z}<extra></extra>",
         colorscale='RdBu'
    )
)

fig.update_yaxes(showticklabels=False)

fig.update_layout(
    xaxis_title="Experimental group",
    yaxis_title="homer.closestTSS_perGene.tsv",
    height=2000,
    width=1000
)
fig.show()
In [ ]: